Graph connection Laplacian and random matrices with random blocks

10/01/2013
by   Noureddine El Karoui, et al.
0

Graph connection Laplacian (GCL) is a modern data analysis technique that is starting to be applied for the analysis of high dimensional and massive datasets. Motivated by this technique, we study matrices that are akin to the ones appearing in the null case of GCL, i.e the case where there is no structure in the dataset under investigation. Developing this understanding is important in making sense of the output of the algorithms based on GCL. We hence develop a theory explaining the behavior of the spectral distribution of a large class of random matrices, in particular random matrices with random block entries of fixed size. Part of the theory covers the case where there is significant dependence between the blocks. Numerical work shows that the agreement between our theoretical predictions and numerical simulations is generally very good.

READ FULL TEXT

page 26

page 27

research
05/23/2014

Connection graph Laplacian methods can be made robust to noise

Recently, several data analytic techniques based on connection graph lap...
research
09/11/2019

Unified ℓ_2→∞ Eigenspace Perturbation Theory for Symmetric Random Matrices

Modern applications in statistics, computer science and network science ...
research
11/20/2020

On Random Matrices Arising in Deep Neural Networks: General I.I.D. Case

We study the distribution of singular values of product of random matric...
research
08/10/2020

Tracy-Widom distribution for the edge eigenvalues of Gram type random matrices

Large dimensional Gram type matrices are common objects in high-dimensio...
research
03/01/2022

Riemannian statistics meets random matrix theory: towards learning from high-dimensional covariance matrices

Riemannian Gaussian distributions were initially introduced as basic bui...
research
12/23/2022

Equi-isoclinic subspaces, covers of the complete graph, and complex conference matrices

In 1992, Godsil and Hensel published a ground-breaking study of distance...
research
10/04/2021

Unraveling the graph structure of tabular datasets through Bayesian and spectral analysis

In the big-data age tabular datasets are being generated and analyzed ev...

Please sign up or login with your details

Forgot password? Click here to reset